---
title: "Coronavirus in Australia"
author: "Jason Everett"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    # social: ["facebook", "twitter", "linkedin"]
    source_code: embed
    vertical_layout: fill
---

```{r setup, include=FALSE}
#------------------ Packages ------------------
library(flexdashboard)
library(tidyverse)
# install.packages("devtools")
# devtools::install_github("RamiKrispin/coronavirus", force = TRUE)
library(coronavirus)
# data(coronavirus)
# update_datasets()
# View(coronavirus)
# max(coronavirus$date)

#----------------------------------------------------
# Pulling the coronvirus data from John Hopkins repo
# https://github.com/CSSEGISandData/COVID-19
#----------------------------------------------------
# Setting functions
# `%>%` <- magrittr::`%>%`
#----------------------------------------------------
#------------ Pulling confirmed cases------------
conf_url <- "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv"
raw_conf <- read.csv(file = conf_url,
                     stringsAsFactors = FALSE)

lapply(1:ncol(raw_conf), function(i){
  if(all(is.na(raw_conf[, i]))){
    raw_conf <<- raw_conf[, -i]
    return(print(paste("Column", names(raw_conf)[i], "is missing", sep = " ")))
  } else {
    return(NULL)
  }
})

# Transforming the data from wide to long
# Creating new data frame
df_conf <- raw_conf[, 1:4]

for(i in 5:ncol(raw_conf)){
  
  raw_conf[,i] <- as.integer(raw_conf[,i])
  # raw_conf[,i] <- ifelse(is.na(raw_conf[, i]), 0 , raw_conf[, i])
  print(names(raw_conf)[i])
  
  if(i == 5){
    df_conf[[names(raw_conf)[i]]] <- raw_conf[, i]
  } else {
    df_conf[[names(raw_conf)[i]]] <- raw_conf[, i] - raw_conf[, i - 1]
  }
  
  
}


df_conf1 <- df_conf %>% 
  tidyr::pivot_longer(cols = dplyr::starts_with("X"),
                      names_to = "date_temp",
                      values_to = "cases_temp")

# Parsing the date
df_conf1$month <- sub("X", "",
                      strsplit(df_conf1$date_temp, split = "\\.") %>%
                        purrr::map_chr(~.x[1]) )

df_conf1$day <- strsplit(df_conf1$date_temp, split = "\\.") %>%
  purrr::map_chr(~.x[2])


df_conf1$date <- as.Date(paste("2020", df_conf1$month, df_conf1$day, sep = "-"))

# Aggregate the data to daily
df_conf2 <- df_conf1 %>%
  dplyr::group_by(Province.State, Country.Region, Lat, Long, date) %>%
  dplyr::summarise(cases = sum(cases_temp)) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(type = "confirmed",
                Country.Region = trimws(Country.Region),
                Province.State = trimws(Province.State))

head(df_conf2)
tail(df_conf2)
#----------------------------------------------------
# Pulling death cases

death_url <- "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv"
raw_death <- read.csv(file =death_url,
                      stringsAsFactors = FALSE,
                      fill =FALSE)

lapply(1:ncol(raw_death), function(i){
  if(all(is.na(raw_death[, i]))){
    raw_death <<- raw_death[, -i]
    return(print(paste("Column", names(raw_death)[i], "is missing", sep = " ")))
  } else {
    return(NULL)
  }
})

# Transforming the data from wide to long
# Creating new data frame
df_death <- raw_death[, 1:4]

for(i in 5:ncol(raw_death)){
  print(i)
  raw_death[,i] <- as.integer(raw_death[,i])
  raw_death[,i] <- ifelse(is.na(raw_death[, i]), 0 , raw_death[, i])
  
  if(i == 5){
    df_death[[names(raw_death)[i]]] <- raw_death[, i]
  } else {
    df_death[[names(raw_death)[i]]] <- raw_death[, i] - raw_death[, i - 1]
  }
}


df_death1 <-  df_death %>% tidyr::pivot_longer(cols = dplyr::starts_with("X"),
                                               names_to = "date_temp",
                                               values_to = "cases_temp")

# Parsing the date
df_death1$month <- sub("X", "",
                       strsplit(df_death1$date_temp, split = "\\.") %>%
                         purrr::map_chr(~.x[1]) )

df_death1$day <- strsplit(df_death1$date_temp, split = "\\.") %>%
  purrr::map_chr(~.x[2])


df_death1$date <- as.Date(paste("2020", df_death1$month, df_death1$day, sep = "-"))

# Aggregate the data to daily
df_death2 <- df_death1 %>%
  dplyr::group_by(Province.State, Country.Region, Lat, Long, date) %>%
  dplyr::summarise(cases = sum(cases_temp)) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(type = "death",
                Country.Region = trimws(Country.Region),
                Province.State = trimws(Province.State))

#---------------- Aggregate all cases ----------------

#
coronavirus <- dplyr::bind_rows(df_conf2, df_death2) %>%
  as.data.frame()


#------------------ Parameters ------------------
# Set colors
# https://www.w3.org/TR/css-color-3/#svg-color
confirmed_color <- "purple"
active_color <- "#1f77b4"
recovered_color <- "forestgreen"
death_color <- "red"
#------------------ Data ------------------
df <- coronavirus %>%
  # dplyr::filter(date == max(date)) %>%
  dplyr::filter(Country.Region == "Australia") %>%
  dplyr::group_by(Country.Region, type) %>%
  dplyr::summarise(total = sum(cases)) %>%
  tidyr::pivot_wider(
    names_from = type,
    values_from = total
  ) %>%
  # dplyr::mutate(unrecovered = confirmed - ifelse(is.na(recovered), 0, recovered) - ifelse(is.na(death), 0, death)) %>%
  dplyr::mutate(unrecovered = confirmed - ifelse(is.na(death), 0, death)) %>%
  dplyr::arrange(-confirmed) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(country = dplyr::if_else(Country.Region == "United Arab Emirates", "UAE", Country.Region)) %>%
  dplyr::mutate(country = dplyr::if_else(country == "Mainland China", "China", country)) %>%
  dplyr::mutate(country = dplyr::if_else(country == "North Macedonia", "N.Macedonia", country)) %>%
  dplyr::mutate(country = dplyr::if_else(country == "New Zealand", "NewZealand", country)) %>%
  dplyr::mutate(country = trimws(country)) %>%
  dplyr::mutate(country = factor(country, levels = country))

df_daily <- coronavirus %>%
  dplyr::filter(Country.Region == "Australia") %>%
  dplyr::group_by(date, type) %>%
  dplyr::summarise(total = sum(cases, na.rm = TRUE)) %>%
  tidyr::pivot_wider(
    names_from = type,
    values_from = total
  ) %>%
  dplyr::arrange(date) %>%
  dplyr::ungroup() %>%
  #dplyr::mutate(active = confirmed - death - recovered) %>%
  dplyr::mutate(active = confirmed - death) %>%
  dplyr::mutate(
    confirmed_cum = cumsum(confirmed),
    death_cum = cumsum(death),
    # recovered_cum = cumsum(recovered),
    active_cum = cumsum(active)
  )


df1 <- coronavirus %>% dplyr::filter(date == max(date))
```

Summary
=======================================================================

Row {data-width=400}
-----------------------------------------------------------------------

### confirmed {.value-box}

```{r}

valueBox(
  value = paste(format(sum(df$confirmed), big.mark = ","), "", sep = " "),
  caption = "Total confirmed cases",
  icon = "fas fa-user-md",
  color = confirmed_color
)
```
















### death {.value-box}

```{r}

valueBox(
  value = paste(format(sum(df$death, na.rm = TRUE), big.mark = ","), " (",
                round(100 * sum(df$death, na.rm = TRUE) / sum(df$confirmed), 1),
                "%)",
                sep = ""
  ),
  caption = "Death cases (death rate)",
  icon = "fas fa-heart-broken",
  color = death_color
)
```


Row
-----------------------------------------------------------------------

### **Daily cumulative cases by type** (Australia only)

```{r}
plotly::plot_ly(data = df_daily) %>%
  plotly::add_trace(
    x = ~date,
    # y = ~active_cum,
    y = ~confirmed_cum,
    type = "scatter",
    mode = "lines+markers",
    # name = "Active",
    name = "Confirmed",
    line = list(color = active_color),
    marker = list(color = active_color)
  ) %>%
  plotly::add_trace(
    x = ~date,
    y = ~death_cum,
    type = "scatter",
    mode = "lines+markers",
    name = "Death",
    line = list(color = death_color),
    marker = list(color = death_color)
  ) %>%
  plotly::add_annotations(
    x = as.Date(df_daily$date[df_daily$confirmed>0][1]),
    y = 1,
    text = paste("First case"),
    xref = "x",
    yref = "y",
    arrowhead = 5,
    arrowhead = 3,
    arrowsize = 1,
    showarrow = TRUE,
    ax = -10,
    ay = -90
  ) %>%
  plotly::add_annotations(
    x = as.Date(df_daily$date[df_daily$death>0][1]),
    y = 3,
    text = paste("First death"),
    xref = "x",
    yref = "y",
    arrowhead = 5,
    arrowhead = 3,
    arrowsize = 1,
    showarrow = TRUE,
    ax = -20,
    ay = -90
  ) %>%
  # plotly::add_annotations(
  #   x = as.Date("2020-03-23"),
  #   y = 100,
  #   text = paste(
  #     "New containment",
  #     "
",
  #     "measures"
  #   ),
  #   xref = "x",
  #   yref = "y",
  #   arrowhead = 5,
#   arrowhead = 3,
#   arrowsize = 1,
#   showarrow = TRUE,
#   ax = -30,
#   ay = -90
# ) %>%
# plotly::add_annotations(
#   x = as.Date("2020-03-25"),
#   y = 10,
#   text = paste(
#     "Stricter containment",
#     "
",
#     "measures"
#   ),
#   xref = "x",
#   yref = "y",
#   arrowhead = 5,
#   arrowhead = 3,
#   arrowsize = 1,
#   showarrow = TRUE,
#   ax = 10,
#   ay = -180
# ) %>%
plotly::layout(
  title = "",
  yaxis = list(title = "Cumulative number of cases"),
  xaxis = list(title = "Date"),
  legend = list(x = 0.1, y = 0.9),
  hovermode = "compare"
)
```

Comparison
=======================================================================


Column {data-width=400}
-------------------------------------


### **Daily new cases**

```{r}
daily_confirmed <- coronavirus %>%
  dplyr::filter(type == "confirmed") %>%
  dplyr::filter(date >= "2020-02-29") %>%
  dplyr::mutate(country = Country.Region) %>%
  dplyr::mutate(country = dplyr::if_else(country == "New Zealand", "NewZealand", country)) %>%
  dplyr::filter(Country.Region == "Australia" |
                  Country.Region == "New Zealand" |
                  Country.Region == "China" |
                  Country.Region == "US" | 
                  Country.Region == "Italy") %>%
  dplyr::group_by(date, country) %>%
  dplyr::summarise(total = sum(cases)) %>%
  dplyr::ungroup() %>%
  tidyr::pivot_wider(names_from = country, values_from = total)

#----------------------------------------
# Plotting the data

daily_confirmed %>%
  plotly::plot_ly() %>%
  plotly::add_trace(
    x = ~date,
    y = ~Australia,
    type = "scatter",
    mode = "lines+markers",
    name = "Australia"
  ) %>%
  plotly::add_trace(
    x = ~date,
    y = ~NewZealand,
    type = "scatter",
    mode = "lines+markers",
    name = "New Zealand"
  ) %>%
  plotly::add_trace(
    x = ~date,
    y = ~China,
    type = "scatter",
    mode = "lines+markers",
    name = "China"
  ) %>%
  plotly::add_trace(
    x = ~date,
    y = ~US,
    type = "scatter",
    mode = "lines+markers",
    name = "US"
  ) %>%
  plotly::add_trace(
    x = ~date,
    y = ~Italy,
    type = "scatter",
    mode = "lines+markers",
    name = "Italy"
  ) %>%
  plotly::layout(
    title = "",
    legend = list(x = 0.1, y = 0.9),
    yaxis = list(title = "Number of new cases"),
    xaxis = list(title = "Date"),
    # paper_bgcolor = "black",
    # plot_bgcolor = "black",
    # font = list(color = 'white'),
    hovermode = "compare",
    margin = list(
      # l = 60,
      # r = 40,
      b = 10,
      t = 10,
      pad = 2
    )
  )
```

### **Cases distribution by type**

```{r daily_summary}
df_region <- coronavirus %>%
  # dplyr::filter(date == max(date)) %>%
  dplyr::filter(Country.Region == "Australia" |
                  Country.Region == "New Zealand" |
                  Country.Region == "China" |
                  Country.Region == "US" | 
                  Country.Region == "Italy") %>%
  dplyr::group_by(Country.Region, type) %>%
  dplyr::summarise(total = sum(cases)) %>%
  tidyr::pivot_wider(
    names_from = type,
    values_from = total
  ) %>%
  # dplyr::mutate(unrecovered = confirmed - ifelse(is.na(recovered), 0, recovered) - ifelse(is.na(death), 0, death)) %>%
  dplyr::mutate(unrecovered = confirmed - ifelse(is.na(death), 0, death)) %>%
  dplyr::arrange(confirmed) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(country = dplyr::if_else(Country.Region == "United Arab Emirates", "UAE", Country.Region)) %>%
  dplyr::mutate(country = dplyr::if_else(country == "Mainland China", "China", country)) %>%
  dplyr::mutate(country = dplyr::if_else(country == "North Macedonia", "N.Macedonia", country)) %>%
  dplyr::mutate(country = trimws(country)) %>%
  dplyr::mutate(country = factor(country, levels = country))

plotly::plot_ly(
  data = df_region,
  x = ~country,
  # y = ~unrecovered,
  y = ~ confirmed,
  # text =  ~ confirmed,
  # textposition = 'auto',
  type = "bar",
  name = "Confirmed",
  marker = list(color = active_color)
) %>%
  plotly::add_trace(
    y = ~death,
    # text =  ~ death,
    # textposition = 'auto',
    name = "Death",
    marker = list(color = death_color)
  ) %>%
  plotly::layout(
    barmode = "stack",
    yaxis = list(title = "Total cases"),
    xaxis = list(title = ""),
    hovermode = "compare",
    margin = list(
      # l = 60,
      # r = 40,
      b = 10,
      t = 10,
      pad = 2
    )
  )
```


Map
=======================================================================

### **World map of cases** (*use + and - icons to zoom in/out*)

```{r}
# map tab added by Art Steinmetz
library(leaflet)
library(leafpop)
library(purrr)
cv_data_for_plot <- coronavirus %>%
  # dplyr::filter(Country.Region == "m") %>%
  dplyr::filter(cases > 0) %>%
  dplyr::group_by(Country.Region, Province.State, Lat, Long, type) %>%
  dplyr::summarise(cases = sum(cases)) %>%
  dplyr::mutate(log_cases = 2 * log(cases)) %>%
  dplyr::ungroup()
cv_data_for_plot.split <- cv_data_for_plot %>% split(cv_data_for_plot$type)
pal <- colorFactor(c("orange", "red", "green"), domain = c("confirmed", "death", "recovered"))
map_object <- leaflet() %>% addProviderTiles(providers$Stamen.Toner)
names(cv_data_for_plot.split) %>%
  purrr::walk(function(df) {
    map_object <<- map_object %>%
      addCircleMarkers(
        data = cv_data_for_plot.split[[df]],
        lng = ~Long, lat = ~Lat,
        #                 label=~as.character(cases),
        color = ~ pal(type),
        stroke = FALSE,
        fillOpacity = 0.8,
        radius = ~log_cases,
        popup = leafpop::popupTable(cv_data_for_plot.split[[df]],
                                    feature.id = FALSE,
                                    row.numbers = FALSE,
                                    zcol = c("type", "cases", "Country.Region", "Province.State")
        ),
        group = df,
        #                 clusterOptions = markerClusterOptions(removeOutsideVisibleBounds = F),
        labelOptions = labelOptions(
          noHide = F,
          direction = "auto"
        )
      )
  })

map_object %>%
  addLayersControl(
    overlayGroups = names(cv_data_for_plot.split),
    options = layersControlOptions(collapsed = FALSE)
  )
```





About
=======================================================================

**The Coronavirus Dashboard: the case of Australia**

This Coronavirus dashboard: the case of Australia provides an overview of the 2019 Novel Coronavirus COVID-19 (2019-nCoV) epidemic for m. This dashboard is built with R using the R Makrdown framework and was adapted from this [dashboard](https://ramikrispin.github.io/coronavirus_dashboard/){target="_blank"} by Rami Krispin.

**Code**

The code behind this dashboard is available on [GitHub](https://github.com/AntoineSoetewey/coronavirus_dashboard){target="_blank"}.

**Data**

The input data for this dashboard is the dataset available from the [`{coronavirus}`](https://github.com/RamiKrispin/coronavirus){target="_blank"} R package. Make sure to download the development version of the package to have the latest data:

```
install.packages("devtools")
devtools::install_github("RamiKrispin/coronavirus")
```

The data and dashboard are refreshed on a daily basis.

The raw data is pulled from the Johns Hopkins University Center for Systems Science and Engineering (JHU CCSE) Coronavirus [repository](https://github.com/RamiKrispin/coronavirus-csv){target="_blank"}.

**Contact**

For any question or feedback, you can [contact me](https://www.statsandr.com/contact/). More information about this dashboard can be found in this [article](https://www.statsandr.com/blog/how-to-create-a-simple-coronavirus-dashboard-specific-to-your-country-in-r/).

**Update**

The data is as of `r format(max(coronavirus$date), "%A %B %d, %Y")` and the dashboard has been updated on `r format(Sys.time(), "%A %B %d, %Y")`.